Phosphonate analogues of arabinose 5-phosphate: Putative ligands for arabinose 5-phosphate isomerases

Article abstract of DOI:10.1002/ejoc.201300887

Metabolically stable arabinose 5-phosphate analogues possessing phosphate mimetic groups at the 5-position were synthesized and evaluated by saturation-transfer-difference (STD) NMR studies for their ability to interact with arabinose 5-phosphate isomeras

Full text of DOI:10.1002/ejoc.201300887

FULL PAPER
DOI: 10.1002/ejoc.201300887
Phosphonate Analogues of Arabinose 5-Phosphate: Putative Ligands for
Arabinose 5-Phosphate Isomerases
Luca Gabrielli,[a] Cristina Airoldi,[a] Paola Sperandeo,[a] Serena Gianera,[a]
Alessandra Polissi,[a] Francesco Nicotra,[a] and Laura Cipolla*[a]
Keywords: Arabinose 5-phosphate / Sugar isomerase / NMR interaction studies / Phosphonate analogues / Phosphorus /
Antibiotics / Biosynthesis / Pseudomonas aeruginosa
Metabolically stable arabinose 5-phosphate analogues pos-
sessing phosphate mimetic groups at the 5-position were
synthesized and evaluated by saturation-transfer-difference
(STD) NMR studies for their ability to interact with arabinose
5-phosphate isomerase. Isosteric methylphosphonate 1 and
(difluoromethyl)phosphonate 3 were recognised and bound
by the catalytic pocket of the enzyme. The synthesized com-
pounds were tested in vivo on E. coli and P. aeruginosa
strains in order to evaluate their antibacterial activity.
zymes and mechanisms involved in Kdo biosynthesis.[8]
These studies are also encouraged by the observation that
a breakdown of Kdo biosynthesis leads to accumulation of
lipid A precursors inside the cell, affecting bacterial viabil-
ity.[9] Moreover, in enterobacterial LPSs, the linkage of the
core region to lipid A always occurs through a Kdo residue.
The Kdo biosynthetic pathway is crucial to conserved (con-
stitutive) intracellular enzymes present in virtually all
Gram-negative bacteria and that are not generally subject
to regulation.[10] These observations suggest that Kdo bio-
synthetic enzymes are excellent therapeutic targets by which
to achieve a broad spectrum of activity against Gram-nega-
tive bacteria.[11]
An underexploited enzyme involved in Kdo biosynthesis
is arabinose 5-phosphate isomerase (API). API catalyses the
reversible isomerization of ribulose 5-phosphate (Ru5P)
into arabinose 5-phosphate (A5P). As a matter of fact, in
several Gram-negative bacteria, API inhibition increases
the susceptibility of the bacteria to antibiotics and, at the
same time, decreases the risk of sepsis by lowering endo-
toxin burdens.[12] Only a few attempts to identify API inhib-
itors have appeared in the literature.[13,14] The most effective
compounds in vitro were described by Woodard and co-
workers in 2011,[14] and their binding mode to API has been
recently characterized in collaboration with our group.[15]
Despite the interesting IC50 in vitro, these molecules were
active in vivo only in millimolar concentrations, suggesting
that the search for more potent inhibitors is still open.
Previous NMR studies of API[16] by our group showed
that the acidic phosphate group is fundamental to enzyme
binding. In particular, we speculate that a salt bridge be-
tween the negatively charged phosphate and a complemen-
tary basic residue in the catalytic pocket is the key interac-
tion between substrate and enzyme.[17] Due to the poor in
Introduction
New opportunistic pathogens have emerged in the last
several decades, whereas older well-known pathogens
thought to have been defeated have re-emerged. In addition,
the onset of multidrug resistant bacterial strains,[1–3] further
exacerbates this scenario demanding continuous identifica-
tion and study of novel antibiotics. Lipopolysaccharide
(LPS), a key constituent of the Gram-negative bacteria
outer membrane, represents a major virulence factor elicit-
ing infection, inflammation and ultimately, septicemia.[4]
For these reasons, the biosynthetic machinery responsible
for LPS production is considered a primary target for the
development of antibacterial and anti-septicemia drugs.[5]
From a chemical perspective, LPS is an essential glyco-
lipid located in the outer membrane of Gram-negative bac-
teria composed of three distinct regions: lipid A, the core
oligosaccharide, and O-antigen regions. The core region is
negatively charged due to the presence of phosphate substit-
uents and/or sugar acids like Kdo and uronic acids. This
negative charge is thought to contribute to the rigidity of
the Gram-negative cell wall through intermolecular cationic
crosslinks.[6] Among the saccharide residues constituting
LPS, 2-keto-3-deoxy-d-manno-octulosonic acid (Kdo) is an
essential component of LPSs of all Gram-negative bacteria
known to date.[7] The significance and ubiquitous presence
of this acid monosaccharide within LPSs promoted investi-
gations into its biosynthesis and a detailed study of the en-
[a] Dept. of Biotechnology and Biosciences, University of Milano-
Bicocca,
Piazza della Scienza 2, 20126 Milano, Italy
E-mail: laura.cipolla@unimib.it
http://www.unimib.it/go/102/Home/English
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201300887.
7776
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 7776–7784

Phosphonate Analogues of Arabinose 5-Phosphate
vivo stability of the phosphate group toward enzymatic hy- its ability to resist enzymatic and chemical hydrolysis. First,
drolysis,[18] identification of good and metabolically resist- it should be expected to significantly decrease the acidity
ant phosphate mimetics is a key point for the design of of the phosphorus-containing acid function moving from a
potential API inhibitors.
phosphate to a phosphonate moiety. This is attributable to
We present here the design, synthesis and biological the introduction of an electron-donating alkyl group, re-
evaluation of a set of A5P analogues, modified at the 5- sulting in a different dissociation state for the analogue
position by the introduction of phosphate mimetic groups compared to the natural compound at physiological pH.[19]
(Figure 1).
Compound 1 was synthesized by Unger in 1978 and tested
in vitro on isolated Kdo8P synthase.[20] From these experi-
ments it was determined that 1 is a substrate for Kdo8P
synthase; 1 is converted to 3,8,9-trideoxy-9-phosphono-d-
manno-nonulosonate, a trigger for feedback inhibition of
the same enzyme. However, no in vivo or in vitro assays
with API have ever been performed. Thus, we report here
the synthesis of compound 1, by an improved synthetic
methodology, affording the desired analogue in fewer syn-
thetic steps (9 instead of 13) and with higher total yield
(28% instead of 18%) compared to what already existed in
the literature.[20]
Figure 1. Set of A5P analogues with different acidic groups at posi-
tion 5.
To study the effect of a non-isosteric phosphonate the α-
hydroxyphosphonate analogue 2 was also synthesized, to-
gether with (difluoromethyl)phosphonate analogue 3. (Di-
fluoromethyl)phosphonates are postulated to be stable bio-
logical mimetics of natural phosphates.[18] In fact, the
–CF2– group is an isoelectronic and isosteric analogue of
the oxygen atom. In the last 15 years a wide range of biolo-
gically active compounds containing the difluoromethyl
group have been synthesized, including, but not limited to,
modified nucleosides, amino acids, carbohydrates and het-
erocycles.[18] Some of these derivatives were found to be po-
tent inhibitors of several enzymes and proved to be impor-
tant mechanistic probes for numerous biological processes
involving phosphates.[18] The geometric properties of the
(difluoromethyl)phosphonates are similar to those of the
phosphate.[21] Moreover, the presence of two fluorine atoms
increases the acidity of the phosphonate (second dissoci-
ation pKa 5.4, phosphate 6.2), whereas the methylphos-
phonate has a second pKa (pKa 7.4), lower than the natural
phosphate.[21] Furthermore, it is notable that, although
fluorine is the most abundant halogen and ranks number
13 among all the elements in the earth’s crust, naturally
occurring organofluorine compounds (compounds contain-
ing one or more C–F bonds) are rare. In fact, among nearly
3200 known naturally occurring organohalogen com-
pounds, only very few (around 13) are organofluorine com-
pounds and all of them are monofluorinated.[22] This obser-
vation prompted us to further investigate this synthetic
fluorinated compound as a potential API inhibitor with an-
tibacterial potential.
The ability of the synthetic compounds 1–3 to interact
with API was assayed by saturation-transfer-difference
(STD) NMR studies using the purified enzyme from Pseu-
domonas aeruginosa. Compounds 1–3 were also assayed for
possible in vivo antibacterial activity.
Results and Discussion
The structural features of A5P analogues 1–3 are envi-
sioned to intervene in Kdo biosynthesis in different ways,
as depicted in Figure 2. These agents may act (i) as competi-
tive API inhibitors, thus blocking the first stage of Kdo
biosynthesis; and (ii) since A5P is the substrate of down-
stream enzyme Kdo8P synthase, they may serve as inhibi-
tors of both API and Kdo8P synthase. Notably, since 1–3
possess a metabolically stable phosphate mimic, they will
not be hydrolyzed by the Kdo8P phosphatase in the third
biosynthetic step, thus blocking this stage of Kdo biosyn-
thesis.
Chemistry
Figure 2. Key biosynthetic steps involved in Kdo production and
possible targets for A5P phosphonate analogues.
The synthesis of 1–3 was performed by starting from d-
arabinose (Scheme 1), which was first treated with tert-
In compound 1 the phosphate group was substituted butyldiphenylsilyl chloride (TBDPSCl) in order to selec-
with the isosteric methylphosphonate group. Several aspects tively protect the primary hydroxy group, affording com-
of this functionalization warrant consideration aside from pound 4 (75% yield), and then with acetone dimethyl acetal
Eur. J. Org. Chem. 2013, 7776–7784
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7777

L. Cipolla et al.
FULL PAPER
in dry dichloromethane (DCM), with camphorsulfonic acid urated phosphonate 9 (90% yield over two steps). Hydro-
(CSA) in catalytic amounts, to protect the hydroxy groups genolysis catalysed by platinum oxide in EtOH cleaved the
in positions 1 and 2 (compound 5, 76%). Benzylation of benzyl group with concomitant reduction of the double
the remaining hydroxy group in position 3 (compound 6, bond generating compound 10 (98% yield). Phosphonate
93%) and subsequent selective cleavage of the silyl ether protecting groups were cleaved by trimethylsilyl bromide
with tetrabutylammonium fluoride (TBAF), gave com- (TMSBr) with dry pyridine in acetonitrile to produce com-
pound 7 (80% yield), the key precursor for A5P analogues pound 11, (94% yield). Final hydrolysis of the isopropylid-
1–3, making the 5-position readily available for introduc- ene group was performed in a D2O solution of 11, directly
1
tion of phosphate mimetic groups.
in the NMR tube, and the deprotection monitored by H
NMR spectroscopy; pH neutralization afforded the methyl-
phosphonate analogue 1 in quantitative yield.
Alternatively, reaction of freshly prepared aldehyde 8
with diethyl phosphite[24] in DCM and triethylamine (TEA)
gave α-hydroxyphosphonate 12 in 89% yield over two steps
(Scheme 3). Subsequent cleavage of the benzyl group in
EtOH and Pd(OH)2 (compound 13, quantitative yield), fol-
lowed by deprotection of the phosphonate ethyl groups
with TMSBr and pyridine in acetonitrile gave compound
14 (60% yield). Hydrolysis of the dimethyl acetal and final
pH neutralization gave α-hydroxyphosphonate analogue 2
in quantitative yield.
Scheme 1. Synthesis of the key intermediate 7 for the synthesis of
5-modified A5P analogues.
Methylphosphonate analogue
1
was synthesized
(Scheme 2) by oxidation of compound 7 with Dess–Martin
reagent (DMR) in DCM, affording corresponding aldehyde
8; subsequent treatment with diethyl methylenebis(phos-
phonate)[23] and sodium hydride in dry benzene gave unsat-
Scheme 3. Synthesis of α-hydroxyphosphonate 2.
Furthermore, key intermediate 7 was treated with triflic
anhydride (Tf2O) and 2,6-di-tert-butyl-4-methylpyridine
(DTBMP), to generate triflate intermediate 15 (Scheme 4);
DTBMP was used as a sterically hindered base since pyr-
idine is capable of triflate displacement.[25] Triflate 15 was
immediately treated with diethyl (difluoromethyl)phos-
phonate[26] using lithium diisopropylamide as base, in the
presence of hexamethylphosphoramide (HMPA) in dry
tetrahydrofuran (THF) to afford compound 16 in 30% yield
over the two steps. Subsequent cleavage of the benzyl group
in EtOH and Pd(OH)2 (compound 17, quantitative yield)
and the ethyl groups with TMSBr and pyridine in aceto-
nitrile afforded compound 18 (85% yield). Isopropylidene
cleavage was performed as for compound 11, thus affording
(difluoromethyl)phosphonate analogue 3 in quantitative
yield.
Scheme 2. Synthesis of methylphosphonate 1.
7778
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 7776–7784

Phosphonate Analogues of Arabinose 5-Phosphate
strated for natural substrates.[16] Low-intensity STD can be
observed for the methylene moiety of 1 suggesting involve-
ment, albeit modest, of this group’s protons in the recogni-
tion process. In the case of compound 3 evaluation of the
position 6 contribution to binding was not possible due to
the absence of protons at the 6 position. In addition, STD
competitive binding experiments[16] performed in the pres-
ence of the natural substrates indicated that compounds 1
and 3 compete for enzyme binding with A5P and Ru5P
suggestive of active-site binding by 1 and 3. In particular,
when an STD spectrum was acquired on a mixture contain-
ing API and an equimolar concentration of natural sub-
strates (A5P and Ru5P concentrations were 2.1 mm and
0.9 mm, respectively) and of compounds 1 or 3 (3 mm), the
STD signals of the synthetic ligand disappeared, clearly in-
dicating that our molecules have low affinity for the protein.
As for Ru5P, the substrate with the highest affinity, a KD
of 0.6 mm was calculated in our previous work,[16b] for 1
and 3 we estimate a KD value Ͼ 1 mm, indicating that these
compounds do not yet constitute inhibitors.
Scheme 4. Synthesis of (difluoromethyl)phosphonate analogue 3.
NMR Experiments
On the contrary, compound 2 did not show any affinity
for the target enzyme, since STD-NMR spectroscopic data
showed no interaction with API (NMR spectroscopic data
not shown). This observation suggests that the non-isosteric
α-hydroxyphosphonate group is not a good mimetic of the
phosphate group.
The binding of 1–3 to API form Pseudomonas aeruginosa
was elucidated by STD-NMR spectroscopy.[16,27] The pres-
1
ence of H NMR signals corresponding to the tested com-
pounds in the STD spectra is non-ambiguous evidence of
enzyme binding. Conversely, the absence of small molecule-
1
derived H NMR signals in STD spectra indicates the ab-
sence of enzyme–small molecule associations.
STD-NMR experiments carried out by using 1–3 in the
presence of target protein revealed that compounds 1 and
Antibacterial Activity Assays
The antibacterial activities of compounds 1–3 were
3 are API ligands (Figure 3). 1H NMR-STD spectra clearly evaluated by using the micro-broth dilution procedure[28]
indicate that, in both cases, positions 1 and 3 are the most using the wild-type Pseudomonas aeruginosa PAO1 strain
important for enzyme interactions as previously demon- and two Escherichia coli K12 strains: the wild type
Figure 3. (a) 1H NMR spectrum of compound 1 (3 mm), 64 scans; (b) STD-NMR spectrum of compound 1 (3 mm), in the presence of
API (30 μm) acquired with 1024 scans, 2 s of saturation times and a saturation frequency of 7.9 ppm; (c) 1H NMR spectrum of 3 (3 mm),
64 scans; (d) STD-NMR spectrum of compound 3 (3 mm), in the presence of API (30 μm) acquired with 1024 scans, 2 s of saturation
times and a saturation frequency of 7.9 ppm. All samples were dissolved in phosphate buffer saline (PBS), pH = 7.4, at 37 °C.
Eur. J. Org. Chem. 2013, 7776–7784
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7779

L. Cipolla et al.
FULL PAPER
and at 100.57 MHz (13C) with a Varian Mercury instrument.
Chemical shifts are reported in ppm downfield from TMS as an
internal standard; J values are given in Hz. Mass spectra were re-
corded with a System Applied Biosystems MDS SCIEX instrument
(Q TRAP, LC/MS/MS, turbon ion spray) or with a System Applied
Biosystem MDS SCIEX instrument (Q STAR elite nanospray).
ESI full MS data were recorded by using a Thermo LTQ instru-
ment by direct inlet.
BW25113 strain[29] and the AS19 mutant strain.[30] The
AS19 mutant strain is characterized by a severely depleted
LPS layer resulting in an outer membrane that is abnor-
mally permeable to antibiotics. This mutant enables one to
understand the extent to which inhibitory effects of drugs
or drug candidates are limited by membrane permeability.
In addition, since A5P uptake is allowed in E. coli by the
hexose phosphate transport system (uhp), inducible by
glucose 6-phosphate G6P,[31] the antibacterial activities of
1–3 against the BW25113 strain was evaluated both in the
presence and absence of 10 mm G6P.
Compounds 1–3 were assayed with the different strains
at increasing concentrations up to 1 mm. As expected, com-
pound 2, unable to interact with the target enzyme in vitro,
as determined by NMR studies, failed to show any activity
against the selected strains in vivo. Unfortunately, the same
trend was observed also with compounds 1 and 3, despite
their affinity for API as indicated by the preceding NMR
studies.
Silyl Ether 4: To a stirred suspension of d-arabinose (1.83 g,
12.24 mmol, 1 equiv.) in dry pyridine (30 mL), TBDPSC1
(3.83 mL, 14.9 mmol, 1.2 equiv.) was added dropwise at 0 °C under
argon. After 10 min, the reaction mixture was stirred at 4 °C over-
night. EtOH was added to quench the reaction, and the mixture
was stirred for 15 min. The mixture was concentrated under re-
duced pressure. The product was purified by flash column
chromatography (5:5, PE/EtOAc) to afford 4 (3.547 g, 75% yield)
as a yellow oil (mixture of α- and β-anomers). NMR and MS data
are in agreement with those previously reported in the literature.[32]
Dimethyl Acetal 5: To a stirred solution of 4 (3.216 g, 8.278 mmol,
1 equiv.) in dry CH2Cl2 (190 mL), 2,2-dimethoxypropane (2.18 mL,
17.79 mmol, 2 equiv.) and camphorsulfonic acid (0.3 g,
0.828 mmol, 0.1 equiv.) were added at 4 °C under argon. The reac-
tion mixture was stirred at 4 °C overnight. After 14 h, Et3N was
added to neutralize the reaction mixture, which was stirred for
15 min. The mixture was concentrated under reduced pressure. The
product was purified by flash column chromatography (8:2, PE/
EtOAc) to give 5 (2.69 g, 76% yield) as a yellow oil. NMR and
MS data are in agreement with those previously reported in the
literature.[28]
Conclusions
We presented here the design, synthesis and biological
evaluation of a set of A5P analogues, modified at the acidic
group in position 5 with metabolically stable phosphate mi-
metic groups. The synthesis of isosteric phosphonate ana-
logue 1 was performed by using a new and improved syn-
thetic methodology. In addition, for the first time, α-hy-
droxyphosphonate 2 and (difluoromethyl)phosphonate an-
alogue 3 were synthesized.
STD-NMR studies revealed that compounds 1 and 3 are
API ligands, binding the enzyme active site with low affin-
ity. Unfortunately, compounds 1–3 did not show any anti-
bacterial activity against assorted Pseudomonas aeruginosa
and Escherichia coli strains. These results suggest that the
phosphate mimetic groups phosphonate and (difluo-
romethyl)phosphonate are recognised by the enzyme but
that they do not permit tight binding. These data indicate
that the design of potent API and Kdo biosynthesis inhibi-
tors warrants the continued search for effective substrate
analogues.
Benzyl Ether 6: To a stirred solution of 5 (1.250 g, 2.919 mmol,
1 equiv.) in dry THF (20 mL), benzyl bromide (1.84 mL,
11.676 mmol, 4 equiv.) and sodium hydride 60% (584 mg,
14.595 mmol, 5 equiv.) were slowly added portionwise. The reac-
tion mixture was stirred for 3 h, and then the mixture was concen-
trated under reduced pressure. The product was purified by flash
column chromatography (100% PE Ǟ 9:1, PE/EtOAc) to give 6
(1.40 g, 93% yield) as a yellowish oil. 1H NMR (400 MHz, CDCl3):
δ = 7.62–7.54 (m, 5 H, H Ph), 7.38–7.17 (m, 10 H, H Ph), 5.82 (d,
J1,2 = 4.1 Hz, 1 H, H1), 4.59 (d, J2,1 = 4.1 Hz, 1 H, H2), 4.54 (d,
J = 1.5 Hz, 2 H, CH2 Bn), 4.21–4.11 (m, 2 H, H3, H4), 3.80–3.67
(m, 2 H, H5), 1.27 (s, 3 H, Me), 1.22 (s, 3 H, Me), 0.97 (s, 9 H,
tBu) ppm. 13C NMR (101 MHz, CDCl3): δ = 137.77 (1 C, Cquat
C
Ph), 135.93 (1 C, C Ph), 133.50 (1 C, Cquat C Ph), 133.38 (1 C,
Cquat C Ph), 130.09 (1 C, C Ph), 130.03 (1 C, C Ph), 128.83 (1 C,
C Ph), 128.18 (1 C, C Ph), 128.07 (1 C, C Ph), 128.04 (1 C, C Ph),
128.02 (1 C, C Ph), 112.77 (1 C, Cquat), 106.07 (1 C, C1), 85.52 (1
C, C2), 85.43 (1 C, C4), 189 83.15 (1 C, C3), 71.99 (1 C, CH2 Bn),
63.69 (1 C, C5), 27.28 (1 C, Me), 27.14 (3 C, Me tBu), 26.45 (1 C,
Me), 19.52 (1 C, Cquat tBu) ppm. MS: m/z = 541.4 [M + Na]+,
557.4 [M + K]+.
Experimental Section
General Methods: Solvents were dried with molecular sieves for at
least 24 h prior to use when required. When dry conditions were
required, the reactions were performed under Ar or N2. Thin-layer
chromatography (TLC) was performed on silica gel 60F254 coated
glass plates (Merck) with UV detection when possible, charring
with a concd. H2SO4/EtOH/H2O solution (10:45:45 v/v/v), or with
Alcohol 7: To a stirred solution of 6 (1.29 g, 2.49 mmol, 1 equiv.)
in dry THF (20 mL), tetrabutylammonium fluoride (1 m THF solu-
tion, 7.47 mL, 7.47 mmol, 3 equiv.) was added. The reaction mix-
ture was stirred for 1 h and then concentrated under reduced pres-
sure. The product was purified by flash column chromatography
(PE/EtOAc, 7:3 Ǟ 6:4) to give 7 (560 mg, 80% yield) as a white
a solution of (NH4)6Mo7O24 (21 g), Ce(SO4)2 (1 g), concd. H2SO4 solid. 1H NMR (400 MHz, CDCl3): δ = 7.42–7.03 (m, 5 H, H Ph),
(31 mL) in water (500 mL) and then heating to 110 °C for 5 min.
Flash column chromatography was performed by using silica gel
230–400 mesh (Merck). Platinum oxide was filtered by Acrodis®
Premium 25 mm Syringe Filter, with 0.45 µm Nylon membrane.
Routine 1H and 13C NMR spectra were recorded at 400 MHz (1H)
5.83 (d, J1,2 = 3.9 Hz, 1 H, H1), 4.60 (d, J2,1 = 3.9 Hz, 1 H, H2),
4.58–4.45 (m, 2 H, CH2 Bn), 4.15–4.08 (m, 1 H, H4), 3.89 (d, J =
2.5 Hz, 1 H, H3), 3.72–3.59 (m, 2 H, H5), 2.71 (br. s, 1 H, OH),
1.44 (s, 3 H, Me), 1.25 (s, 3 H, Me) ppm. 13C NMR (101 MHz,
CDCl3): δ = 137.39 (1 C, Cquat C Ph), 128.73 (1 C, C Ph), 128.20
7780
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 7776–7784

Phosphonate Analogues of Arabinose 5-Phosphate
(1 C, C Ph), 127.98 (1 C, C Ph), 113.06 (1 C, Cquat), 105.78 (1 C,
NH4OH (3.4 mL) while stirring. The ice bath was removed and the
C1), 85.79 (1 C, C4), 85.38 (1 C, C2), 82.92 (1 C, C3), 72.01 (1 C, solvent evaporated under reduced pressure. The product was puri-
CH2 Bn), 62.79 (1 C, C5), 27.29 (1 C, Me), 26.52 (1 C, Me) ppm.
fied by flash column chromatography (2-propanol/33% aq.
NH4OH, 6:4) to give 11 (98 mg, 94% yield) as a white solid. 1H
NMR (400 MHz, CD3OD): δ = 5.73 (d, J1,2 = 3.8 Hz, 1 H, H1),
4.38 (d, J2,1 = 3.8 Hz, 1 H), 3.87–3.82 (m, 1 H, H3), 3.81–3.72 (m,
1 H, H4), 1.85–1.66 (m, 2 H, H5), 1.41 (s, 3 H, Me), 1.37–1.26 (m,
2 H, H6), 1.22 (s, 3 H, Me) ppm. 13C NMR (101 MHz, CD3OD):
δ = 113.57 (1 C, Cquat), 106.76 (1 C, C1), 89.60 (d, 3JC,P = 17.1 Hz,
MS: m/z = 303.4 [M + Na]+, 319.3 [M + K]+, 583.3 [2M + Na]+.
Aldehyde 8: To a stirred solution of 7 (100 mg, 0.357 mmol,
1 equiv.) in dry DCM (4 mL), Dess–Martin reagent (181 mg,
0.428 mmol, 1.2 equiv.) was added. The reaction mixture was
stirred for 15 min and then neutralized with aq. satd. NaHCO3
solution and aq. satd. Na2S2O3 solution. The mixture was extracted
with DCM; the organic phase was dried with Na2SO4, filtered and
concentrated under reduced pressure. The product was quickly fil-
tered through a silica column (PE/EtOAc, 7:3 + 1% TEA) to give
8 (98 mg) as a light yellow oil that was immediately used for the
next reaction without any characterization.
1 C, C4), 88.49 (1 C, C2), 78.97 (1 C, C3), 29.22 (d, 2JC,P = 3.8 Hz,
1
1 C, C5), 27.32 (1 C, Me), 26.84 (1 C, Me), 26.16 (d, JC,P
=
136.7 Hz, 1 C, C6) ppm. 31P NMR (162 MHz, CD3OD): δ = 24.96
(1 P) ppm. MS: m/z = 267.1 [M – H]–, 535.3 [2 M – H]–.
Deprotected Phosphonate 1: Compound 11 (40 mg, 0.157 mmol)
was dissolved in D2O (0.7 mL), and the solution was transferred
into an NMR tube. The reaction was monitored by 1H NMR spec-
troscopy, and the solution was heated at 90 °C for 3 d. Then the
mixture was neutralized with 5% aq. NaOH, and the solution was
freeze-dried to give 1 in quantitative yield as a mixture of anomers
Diethyl Phosphonate 9: To a solution of NaH 60% (29 mg,
0.714 mmol, 2 equiv.) in dry benzene (5 mL), diethyl methylenebi-
s(phosphonate) (222 µL, 0.983 mmol, 2.5 equiv.) was added, and
then the mixture was stirred at room temperature for 30 min. The
freshly prepared aldehyde 8 (98 mg) was dissolved in dry benzene
(1.5 mL) and added by syringe to the phosphonate solution. The
reaction mixture was stirred at room temperature for 30 min, then
the solvent was evaporated, and the resulting product was extracted
with EtOAc; the organic phase was dried with Na2SO4, filtered,
and concentrated under reduced pressure. The product was purified
by flash column chromatography (PE/EtOAc, 3.5:6.5) to give 9
1
(α/β = 1.5). H NMR (400 MHz, D2O): δ = 5.28 (d, J1,2 = 4.6 Hz,
1 H, H1β), 5.25 (d, J1,2 = 1.8 Hz, 1 H, H1α), 4.12–3.96 (m, 4 H, 2
H2, H4, H3), 3.91–3.85 (m, 1 H, H3), 3.76 (dd, J = 13.0, 7.1 Hz,
1 H, H4), 1.97–1.73 (m, 2 H, H5), 1.67–1.40 (m, 2 H, H6) ppm.
13C NMR (101 MHz, D2O): δ = 100.60 (1 C, C1α), 94.64 (1 C,
C1β), 83.46 (3JC,P = 17.0 Hz, 1 C, d, C4α), 81.41 (3JC,P = 18.3 Hz,
1 C, d, C4β), 81.63, 78.88, 77.19, 76.09 (4 C, 2 C2, 2 C3), 28.98,
27.62 (2 C, C5), 25.16, 23.84 (2 C, C6) ppm. 31P NMR (162 MHz,
D2O): δ = 26.53, 26.31 (2 P) ppm. MS: m/z = 227.1 [M – H]–.
1
(132 mg, 90% yield over two steps) as a light yellow oil. H NMR
(400 MHz, CDCl3): δ = 7.37–7.23 (m, 5 H, H Ph), 6.86–6.73 (m, 1
H, H6), 6.10–6.95 (m, 1 H, H5), 5.94 (d, J1,2 = 3.8 Hz, 1 H), 4.69–
4.50 (m, 4 H, H2, H4, CH2 Bn), 4.11–3.98 (m, 4 H, 2CH2 OEt),
3.95 (d, J = 2.5 Hz, 1 H, H3), 1.45 (s, 3 H, Me), 1.32–1.21 (m, 9
H, Me, 2CH3 OEt) ppm. 13C NMR (101 MHz, CDCl3): δ = 149.13
Diethyl α-Hydroxyphosphonate 12: To a stirred solution of freshly
prepared aldehyde 11 (600 mg) in dry DCM (20 mL), dry TEA
(1280 µL, 10.7 mmol, 5 equiv.) and diethyl phosphite (828 µL,
6.42 mmol, 3 equiv.) were added. The reaction mixture was stirred
for 15 h, then the solvent was evaporated, and the resulting product
was purified by flash column chromatography (PE/EtOAc, 1:1 Ǟ
100% EtOAc) to give 12 (796 mg, 89% yield in two steps) as a light
yellow oil. 1H NMR (400 MHz, CDCl3): δ = 7.41–7.24 (m, 5 H, H
Ph), 5.90 (d, J1,2 = 3.0 Hz, 1 H, H1), 4.69 (d, J = 11.4 Hz, 1 H,
HЈBn), 4.61 (d, J2,1 = 3.0 Hz, 1 H, H2), 4.57 (d, J = 11.4 Hz, 1 H,
HЈЈBn), 4.51–4.44 (m, 1 H, H4), 4.39 (s, 1 H, H3), 4.23–4.12 (m, 4
H, 2CH2 Et), 4.11–4.03 (m, 1 H, H5), 1.53 (s, 3 H, Me), 1.36–1.22
(m, 9 H, Me, 2CH3 Et) ppm. 13C NMR (101 MHz, CDCl3): δ =
137.54 (1 C, Cquat Ph), 128.47 (1 C, C Ph), 128.36 (1 C, C Ph),
127.91 (1 C, C Ph), 127.77 (1 C, C Ph), 112.62 (1 C, Cquat), 106.15
(1 C, C1), 85.16 (1 C, C4), 84.81 (1 C, C2), 82.35 (1 C, C3), 71.59
(1 C, CH2 Bn), 67.83 (J = 161.1 Hz, 1 C, d, C5), 63.09 (d, J =
6.8 Hz, 1 C, CH2 Et), 62.75 (d, J = 6.8 Hz, 1 C, CH2 Et), 26.74,
25.92 (2 C, Me), 16.44 (2 C, CH3 Et) ppm. 31P NMR (162 MHz,
CDCl3): δ = 21.85 (1 P) ppm. MS: m/z = 439.3 [M + Na]+, 855.4
[2 M + Na]+.
2
(d, JC,P = 5.7 Hz, 1 C, C5), 136.81 (1 C, Cquat Ph), 128.63 (1 C, C
1
Ph), 128.17 (1 C, C Ph), 127.83 (1 C, C Ph), 117.60 (d, JC,P
=
188.2 Hz, 1 C, C6), 113.07 (1 C, Cquat), 106.21 (1 C, C1), 85.52,
84.76, 84.50 (C2, C3, C4), 72.01 (1 C, CH2 Bn), 61.75, (2 C, CH2
OEt), 26.53 (1 C, Me), 26.23 (1 C, Me), 16.34, (2 C, CH3 OEt)
ppm. 31P NMR (162 MHz, CDCl3): δ = 17.78 (1 P) ppm. MS: m/z
= 413.3 [M + H]+, 435.4 [M + Na]+, 825.4 [2 M + H]+, 847.4 [2
M + Na]+.
Alcohol 10: Compound 9 (233 mg, 0.212 mmol) was dissolved in
EtOH (6 mL). Palladium hydroxide (200 mg) was added, the mix-
ture was stirred under vacuum in order to remove the residual at-
mosphere, and then H2 was added. The reaction mixture was
stirred at room temperature for 24 h, and then the catalyst was
filtered off and washed with EtOH. The solvent was evaporated to
give 10 as colourless oil (180 mg, 98% yield). 1H NMR (400 MHz,
CDCl3): δ = 5.85 (d, J1,2 = 3.9 Hz, 1 H, H1), 4.54 (br. s, 1 H, OH),
4.50 (d, J2,1 = 3.9 Hz, 1 H, H2), 4.10–3.98 (m, 5 H, H3, 2CH2
OEt), 3.98–3.91 (m, 1 H, H4), 2.08–1.66 (m, 4 H, H5, H6), 1.47 (s,
3 H, Me), 1.30–1.22 (m, 9 H, Me, 2CH3 OEt) ppm. 13C NMR
(101 MHz, CDCl3): δ = 112.28 (1 C, Cquat), 105.74 (1 C, C1), 87.73
Alcohol 13: Compound 12 (375 mg, 0.900 mmol) was dissolved in
EtOH (9 mL). Palladium hydroxide (200 mg) was added, the mix-
ture was stirred under vacuum in order to remove the residual at-
mosphere, and then H2 was added. The reaction mixture was
stirred at room temperature for 24 h, and then the catalyst was
filtered off and washed with EtOH. The solvent was evaporated to
give 13 as a colourless oil (290 mg, 99% yield). 1H NMR
(400 MHz, CD3OD): δ = 5.90 (d, J = 3.5 Hz, 1 H, H1), 4.52 (d, J
= 3.5 Hz, 1 H, H2), 4.41 (s, 1 H, H3), 4.24–4.13 (m, 5 H, H5, 2CH2
Et), 4.10 (s, 1 H, H4), 1.50 (s, 3 H, Me), 1.37–1.26 (m, 9 H, CH3
Et, Me) ppm. 13C NMR (101 MHz, CD3OD): δ = 112.13 (1 C,
Cquat), 106.33 (1 C, C1), 87.19 (1 C, C4), 86.36 (1 C, C2), 74.93 (1
C, C3), 66.87 (J = 165.8 Hz, 1 C, d, C5), 62.60, 62.52 (2 C, CH2
3
(d, JC,P = 16.2 Hz, 1 C, C4), 87.09 (1 C, C2), 77.97 (1 C, C3),
2
61.84 (2 C, CH2 OEt), 26.85 (1 C, Me), 26.66 (d, JC,P = 4.5 Hz, 1
1
C, C5), 26.11 (1 C, Me), 22.06 (d, JC,P = 141.8 Hz, 1 C, C6),
16.44 (d, 3JC,P = 6.3 Hz, 2 C, CH3 OEt) ppm. 31P NMR (162 MHz,
CDCl3): δ = 32.13 (1 P) ppm. MS: m/z = 325.3 [M + H]+, 347.2
[M + Na]+, 363.3 [M + K]+, 671.3 [2 M + Na]+.
Phosphonate 11: To a stirred solution of compound 10 (126 mg,
0.389 mmol) in dry acetonitrile (5 mL), dry pyridine (315 µL,
3.885 mmol, 10 equiv.) and TMSBr (513 µL, 3.885 mmol, 10 equiv.)
were added at room temperature. After 30 min, the solution was
cooled to 0 °C and the reaction quenched by addition of 33% aq.
Eur. J. Org. Chem. 2013, 7776–7784
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7781

L. Cipolla et al.
FULL PAPER
Et), 25.54 (1 C, Me), 24.55 (1 C, Me), 15.31 (2 C, CH3 Et) ppm. CH2 Bn), 4.35–4.20 (m, 4 H, CH2 Et), 4.03 (s, 1 H, H3), 2.69–2.45
31P NMR (162 MHz, CDCl3): δ = 24.20 (1 P) ppm. MS: m/z =
(m, 2 H, H5), 1.52 (s, 3 H, Me), 1.42–1.34 (m, 6 H, CH3 Et), 1.32
(s, 3 H, Me) ppm. 13C NMR (101 MHz, CDCl3): δ = 137.25 (1 C,
Cquat Ph), 128.52 (1 C, C Ph), 128.46 (1 C, C Ph), 128.03 (1 C, C
Ph), 127.98 (1 C, C Ph), 127.86 (1 C, C Ph), 112.49 (1 C, Cquat),
106.05 (1 C, C1), 85.49 (1 C, C3), 84.64 (1 C, C2), 78.68 (1 C, C4),
71.82 (1 C, CH2 Bn), 64.60, 64.55 (2 C, CH2 Et), 26.75 (1 C, Me),
25.95 (1 C, Me), 22.92 (d, J = 27.1 Hz, 1 C, C5), 16.45 (2 C, CH3
Et) ppm. 31P NMR (162 MHz, CDCl3): δ = 6.32 (t, JP,F = 106.8 Hz,
1 P) ppm. 19F NMR (376 MHz, CDCl3) δ = –111.77 (ddd, J =
107.4, 25.3, 15.4 Hz, 1 F), –112.24 (ddd, J = 106.1, 24.2, 16.3 Hz,
1 F) ppm. MS: m/z = 451.3 [M + H]+, 473.3 [M + Na]+, 923.3 [2
M + Na]+.
349.2 [M + Na]+, 375.3 [2 M + Na]+.
α-Hydroxyphosphonate 14: To a stirred solution of compound 13
(242 mg, 0.742 mmol) in dry acetonitrile (7 mL), dry pyridine
(600 µL, 7.42 mmol, 10 equiv.) and TMSBr (979 µL, 7.42 mmol,
10 equiv.) were added at room temperature. After 50 min, the solu-
tion was cooled to 0 °C and the reaction quenched by adding aq.
NH4OH (33%, 5 mL) whilst stirring. The ice bath was removed
and the solvent evaporated under reduced pressure. The product
was purified by flash column chromatography (2-propanol/33% aq.
1
NH4OH, 6:4) to give 14 (120 mg, 60% yield) as a white solid. H
NMR (400 MHz, CD3OD): δ = 5.78 (d, J = 4.2 Hz, 1 H), 4.60 (br.
s, 1 H, H2), 4.46 (d, J = 3.8 Hz, 1 H, H3), 4.08–3.96 (m, 2 H, H4,
H5), 1.56 (s, 3 H, Me), 1.34 (s, 3 H, Me) ppm. 13C NMR (101 MHz,
CD3OD): δ = 113.31 (1 C, Cquat), 103.99 (1 C, C1), 88.04 (1 C,
Alcohol 17: Compound 16 (40 mg, 0.218 mmol) was dissolved in
EtOH (1 mL). Palladium hydroxide (40 mg) was added, the mixture
was stirred under vacuum in order to remove the residual atmo-
sphere, and then H2 was added. The reaction mixture was stirred
at room temperature for 24 h, and then the catalyst was filtered off
and washed with EtOH. The solvent was evaporated to give 17 as
2
C2), 84.83 (d, JC,P = 13.8 Hz, 1 C, C4), 73.55 (1 C, C3), 67.97 (d,
1JC,P = 149.1 Hz, 1 C, C5), 26.69, 26.14 (2 C, Me) ppm. 31P NMR
(162 MHz, CD3OD): δ = 16.70 (1 P) ppm. MS: m/z = 269.0 [M –
H]–.
1
a colourless oil (32 mg, 98%). H NMR (400 MHz, CD3OD): δ =
5.88 (d, J1,2 = 3.8 Hz, 1 H, H1), 4.52 (d, J2,1 = 3.8 Hz, 1 H, H2),
4.37–4.24 (m, 5 H, H4, 2CH2 Et), 4.15 (s, 1 H, H3), 2.71–2.41 (m,
2 H, H5), 1.49 (s, 3 H, Me), 1.41–1.36 (m, 6 H, 2CH3 Et), 1.30 (s,
3 H, Me) ppm. 13C NMR (101 MHz, CD3OD): δ = 113.27 (1 C,
Cquat), 107.54 (1 C, C1), 87.69 (1 C, C2), 82.58 (1 C, C4), 79.46 (1
C, C3), 66.33, 66.26 (2 C, CH2 Et), 39.31 (1 C, C5), 26.82 (1 C,
Me), 25.86 (1 C, Me), 16.75, 16.70 (2 C, CH3 Et) ppm. 31P NMR
(162 MHz, CD3OD): δ = 6.41 (t, JP,F = 217.3 Hz, 1 P) ppm. 19F
NMR (376 MHz, CD3OD): δ = –113.26 (app. t, J = 20.4 Hz, 1 F),
–113.55 (app. t, J = 19.4 Hz, 1 F). MS: m/z = 383.2 [M + Na]+,
743.2 [2 M + Na]+.
Deprotected α-Hydroxyphosphonate 2: Compound 14 (40 mg,
0.157 mmol) was dissolved in D2O (0.7 mL), and the solution was
transferred into an NMR tube and heated to 90 °C. The reaction
was monitored by 1H NMR spectroscopy for 5 d. The mixture was
then neutralized by adding 5% aq. NaOH, and the solution was
freeze-dried to give 2 in quantitative yield as a mixture of anomers
of pyranose and furanose forms. Due to the overlapping of signals
of the 4 products, it was not possible to assign clearly each NMR
1
signal. H NMR (400 MHz, D2O): δ = 5.08 (d, J = 4.5 Hz, 1 H),
5.03–4.98 (m, 2 H), 4.89 (s, 1 H), 4.14 (t, J = 8.3 Hz, 2 H), 4.10–
3.93 (m, 4 H), 3.92–3.84 (m, 2 H), 3.80 (dd, J = 14.3, 3.1 Hz, 2 H),
3.73 (t, J = 12.7 Hz, 2 H), 3.63 (d, J = 4.3 Hz, 2 H), 3.53 (t, J =
9.7 Hz, 2 H), 3.42 (t, J = 7.3 Hz, 1 H) ppm. 13C NMR (101 MHz,
D2O): δ = 100.17, 93.76, 93.73, 92.75, 92.63, 81.45, 75.53, 72.34,
71.23, 70.28, 70.01, 68.36, 68.26, 65.74 ppm. 31P NMR (162 MHz,
D2O): δ = 16.94 (1 P), 16.59 (1 P), 16.30 (1 P), 14.77 (1 P) ppm.
MS: m/z = 229.1 [M – H]–.
(Difluoromethyl)phosphonate 18: To a stirred solution of compound
17 (18 mg, 0.05 mmol) in dry acetonitrile (1 mL), dry pyridine
(130 µL, 1.6 mmol, 32 equiv.) and TMSBr (172 µL, 1.3 mmol,
26 equiv.) were added at room temperature. After 10 h, the solution
was cooled to 0 °C and the reaction quenched with pyridine/H2O
(1:1, 4 mL) while stirring. The ice bath was removed, and the sol-
vent was evaporated under reduced pressure. The product was puri-
fied by flash column chromatography (2-propanol/33% aq.
NH4OH, 6:4) to give 18 (13 mg, 85% yield) as a white solid. 1H
NMR (400 MHz, CD3OD): δ = 5.85 (d, J1,2 = 3.9 Hz, 1 H, H1),
4.55 (d, J2,1 = 3.9 Hz, 1 H, H2), 4.34 (t, J = 6.6 Hz, 1 H, H4), 4.16
(s, 1 H, H3), 2.46–2.25 (m, 2 H, H5), 1.38 (s, 3 H, Me), 1.18 (s, 3
H, Me) ppm. 13C NMR (101 MHz, CD3OD): δ = 119.48 (d, JC,F
= 219.5 Hz, 1 C, C6), 113.47 (1 C, Cquat), 107.00 (1 C, C1), 88.24
(1 C, C2), 83.98 (1 C, C4), 79.60 (1 C, C3), 40.28 (1 C, C5), 27.16
(1 C, Me), 26.21 (1 C, Me) ppm. 31P NMR (162 MHz, CD3OD):
δ = 4.08 (t, JP,F = 89.8 Hz, 1 P) ppm. 19F NMR (376 MHz,
CD3OD): δ = –113.34 (app. t, J = 20.1 Hz, 1 F), –113.59 (app. t, J
= 20.0 Hz, 1 F) ppm. MS: m/z = 303.2 [M – H]–.
Triflate 15: To a stirred solution of 7 (110 mg, 0.389 mmol,
1 equiv.) in dry DCM (4 mL), DTBMP (200 mg, 0.975 mmol,
2.5 equiv.) was added. The solution was cooled to –75 °C, and
triflic anhydride (132 µL, 0.785 mmol, 2 equiv.) was added by sy-
ringe; the solution was warmed 0 °C, and after 30 min the reaction
mixture was directly loaded on a silica gel column and quickly
eluted (PE/EtOAc, 6:4) to give 15. The compound was used imme-
diately for the next step because of its lability.
Diethyl (Difluoromethyl)phosphonate 16: A 2 m solution of LDA in
THF (1.17 mL, 2.34 mmol, 6 equiv.) was diluted in dry THF
(2 mL) and cooled to –78 °C. HMPA (407 µL, 2.34 mmol, 6 equiv.)
and diethyl (difluoromethyl)phosphonate (367 µL, 2.34 mmol,
6 equiv.) were dissolved in dry THF (1 mL), and the solution was
cooled to –78 °C and then slowly transferred by cannula to the cold
LDA solution. The mixture was stirred at –78 °C for 20 min, then
a dry THF (2 mL) solution of the freshly prepared triflate 15 was
cooled to –78 °C and transferred slowly via cannula to the cold
LiCF2PO(OEt)2 solution. The mixture was stirred at –78 °C for
25 min, then it was quenched by adding aq. satd. NH4Cl solution.
The product was extracted with EtOAc, the organic phase dried
with Na2SO4, filtered, and concentrated under reduced pressure.
The product was purified by flash column chromatography (PE/
EtOAc, 7:3) to give 16 (50 mg, 30% yield in two steps) as a light
Deprotected (Difluoromethyl)phosphonate 3: Compound 18 (10 mg,
0.033 mmol) was dissolved in D2O (0.6 mL), and the solution was
transferred into an NMR tube and heated to 90 °C for 1 d. The
reaction was monitored by 1H NMR spectroscopy. The mixture
was then neutralized by addition of 5% aq. NaOH and the solution
freeze-dried to give 3 in quantitative yield. 1H NMR (400 MHz,
D2O): δ = 5.07 (d, J1,2 = 4.3 Hz, 1 H, H1β), 5.06 (d, J1,2 = 1.7 Hz,
1 H, H1α), 4.28–4.20 (m, 1 H, H4α), 3.99–3.92 (m, 1 H, H4β),
3.89–3.80 (m, 3 H, H2α, H2β, H3β), 3.75–3.70 (m, 1 H, H3α), 2.40–
2.07 (m, 2 H, H5) ppm. 13C NMR (101 MHz, D2O): δ = 119.22 (d,
yellow oil. 1H NMR (400 MHz, CDCl3): δ = 7.44–7.18 (m, 5 H, H JC,F = 271.0 Hz, 1 C, C6), 99.43 (1 C, C1α), 93.68 (1 C, C1β), 79.43
Ph), 5.91 (d, J1,2 = 3.8 Hz, 1 H, H1), 4.67–4.58 (m, 4 H, H2, H4, (1 C, C3β), 78.07 (1 C, C3α), 76.33 (1 C, C2β), 75.25 (1 C, C4α),
7782
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 7776–7784

Phosphonate Analogues of Arabinose 5-Phosphate
73.96 (1 C, C2α), 73.32 (1 C, C4β), 36.14 (d, JC,F = 163.6 Hz, 1 C,
C5) ppm. 31P NMR (162 MHz, D2O): δ = 4.26 (t, JP,F = 94.8 Hz,
1 P), 4.19 (t, JP,F = 94.8 Hz, 1 P) ppm. 19F NMR (376 MHz, D2O):
δ = –112.36 to –110.81 (m, 2 F), –114.59, 113.42 (m, 2 F). MS: m/z
= 263.1 [M – H]–.
Acknowledgments
This project was funded by Fondazione Cariplo (Grant no. 2010-
0653), Regione Lombardia “Cooperazione scientifica e tecnologica
internazionale” (Grant 16876 SAL-18), Ministero dell’Università e
della Ricerca (MIUR) – Regione Lombardia (ID 30190679), Fond-
azione Banca del Monte di Lombardia, and Ministero dell’Univer-
sità e della Ricerca (MIUR) under project PRIN2008/24M2HX.
NMR Studies: All of the experiments were recorded with a Varian
Mercury 400 MHz instrument. A basic 1D-STD sequence was used
with the on-resonance frequency of 7.9 ppm and the off-resonance
frequency of 40 ppm. A train of Gaussian-shaped pulses of 50 ms
each was employed, with total saturation times of the protein enve-
lope ranging from 3.0 to 0.3 s. The total saturation time was ad-
justed by the number of shaped pulses. A T1ρ filter of 2 ms was
employed to eliminate the background signals from the protein. All
of the samples were dissolved in deuterated PBS, pH = 7.4 at 37 °C.
Total sample volumes were 550 μL. The on- and off-resonance
spectra were acquired interleaving with the same number of scans.
The STD spectrum was obtained by subtraction of the on-reso-
nance spectrum from the off-resonance spectrum. Subtraction was
performed by phase cycling to minimize artifacts arising from mag-
net and temperature instabilities. Reference experiments with sam-
ples containing only the free compounds tested were performed
under the same experimental conditions to verify true ligand bind-
ing. The effects observed in the presence of the protein were due
to true saturation transfer, because no signals were present in the
STD spectra obtained in the reference experiments, except for resi-
dues from HDO, indicating that artifacts from the subtraction of
compound signals were negligible. Competitive STD experiments
(Supporting Information, Figure S1) were acquired on a mixture
containing API and an equimolar concentration of compound 1 or
compound 3 and natural substrates; compound 1 or 3, A5P and
Ru5P concentrations were 3 mm, 2.1 mm and 0.9 mm, respectively.
[1] C. T. Walsh, G. Wright, Chem. Rev. 2005, 105, 391–394.
[2] N. M. Clark, E. Hershberger, M. J. Zervosc, J. P. Lynch, Curr.
Opin. Crit. Care 2003, 9, 403–412.
[3] G. Taubes, Science 2008, 321, 356–361.
[4] a) A. C. De Castro, M. Parrilli, O. Holst, A. Molinaro, Method.
Enzymol. 2010, 480, 89–115; b) B. A. Silipo, A. Molinaro, Sub-
cell. Biochem. 2010, 53, 69–99.
[5] L. Gabrielli, A. Capitoli, D. Bini, F. Taraballi, C. Lupo, L.
Russo, L. Cipolla, Curr. Drug Targets 2012, 13, 1458–1471.
[6] E. Frirdich, C. Bouwman, E. Vinogradov, C. J. Whitfield, Biol.
Chem. 2005, 280, 27604–27612.
[7] L. Cipolla, L. Gabrielli, D. Bini, L. Russo, N. Shaikh, Nat.
Prod. Rep. 2010, 27, 1618–1629.
[8] L. Cipolla, A. Polissi, C. Airoldi, L. Gabrielli, S. Merlo, F.
Nicotra, Curr. Med. Chem. 2011, 18, 830–852.
[9] C. M. Reynolds, C. R. Raetz, Biochemistry 2009, 48, 9627–
9640.
[10] C. R. Raetz, C. Whitfield, Annu. Rev. Biochem. 2002, 71, 635–
700.
[11] J. A. Yethon, C. Whitfield, Curr. Drug Targets Infect. Disord.
2001, 1, 91–106.
[12] T. C. Meredith, P. Aggarwal, U. Mamat, B. Lindner, R. W.
Woodard, ACS Chem. Biol. 2006, 1, 33–42.
[13] E. C. Bigham, C. E. Gragg, W. R. Hall, J. E. Kelsey, W. R.
Mallory, D. C. Richardson, C. Benedict, P. H. Ray, J. Med.
Chem. 1984, 27, 717–726.
Expression and Purification of Pa-KdsD: The P. aeruginosa KdsD
(Pa-KdsD, PA4457) was expressed by pQE31/Pa-KdsD plasmid
that encodes Pa-KdsD with an N-terminal 6-His tag.[12b] Cultures
of E. coli M15/pREP4 harboring pQE31/Pa-KdsD plasmid were
grown overnight in LD broth[33] supplemented with ampicillin
(100 μg mL–1) and kanamycin (25 μg mL–1), diluted 1:100 in fresh
medium (1 L) and grown up to mid-logarithmic phase (OD600, 0.6)
at 37 °C with shaking. Induction of Pa-KdsD was then carried out
with IPTG (0.5 mm) for 4 h. Cells were harvested, washed with so-
dium phosphate (20 mm, pH = 7.0) and re-suspended in sodium
phosphate (50 mm, pH = 8.0), NaCl (300 mm), imidazole (10 mm)
supplemented with DNase (0, 1 mgmL–1) and (phenylmethane)-
sulfonyl fluoride (1 mm). After treatment with lysozyme
(1 mgmL–1) for 30 min, cells were disrupted by a single cycle
through a Cell Disruptor (One Shot Model by Constant Systems
LTD) at 25000 psi, and unbroken cells were removed by centrifuga-
tion at 39000 g for 30 min. The supernatant was loaded at a flow
rate of 0.5 mL min–1 onto an Ni-NTA Agarose (Qiagen) column
(bed volume 5 mL) pre-equilibrated with sodium phosphate
(50 mm, pH = 8.0) and NaCl (300 mm). The column was washed
with 10 volumes of sodium phosphate (50 mm, pH = 8.0), NaCl
(300 mm) and imidazole (20 mm), and the enzyme was eluted with
a 5 mL stepwise imidazole gradient (100 Ǟ 300 mm) in the same
buffer. Fractions (5 mL) were collected and those displaying the
highest purity (Ͼ 95%, as assessed by SDS-PAGE analysis) were
dialyzed in sodium phosphate (50 mm, pH = 8.0). The isomerase
activity of the protein was assessed as described previously.[16b] Pro-
tein aliquots (600 μg each) were lyophilized and stored at –80 °C.
[14] A. Yep, R. J. Sorenson, M. R. Wilson, H. D. H. Showalter,
S. D. Larsen, P. R. Keller, R. W. Woodard, Bioorg. Med. Chem.
Lett. 2011, 21, 2679–2682.
[15] L. Gabrielli, S. Merlo, C. Airoldi, P. Sperandeo, S. Gianera, A.
Polissi, F. Nicotra, T. P. Hollerb, R. W. Woodard, L. Cipolla,
Bioorg. Med. Chem. 2013; DOI: 10.1016/j.bmc.2013.08.012.
[16] a) C. Airoldi, S. Sommaruga, S. Merlo, L. Cipolla, A. Polissi,
F. Nicotra, Chem. Eur. J. 2010, 16, 1897–1902; b) C. Airoldi,
S. Sommaruga, S. Merlo, P. Sperandeo, L. Cipolla, A. Polissi,
F. Nicotra, ChemBioChem 2011, 12, 719–727.
[17] E. C. Bigham, C. E. Gragg, W. R. Hall, J. E. Kelsey, W. R.
Mallory, D. C. Richardson, C. Benedict, P. H. Ray, J. Med.
Chem. 1984, 27, 717–726.
[18] Chunikhin, S. Konstantin, A. A. Kadyrov, P. V. Pasternak, N.
Chkanikov, D. Nikolai, Russ. Chem. Rev. 2010, 79, 371–396.
[19] R. Engel, Chem. Rev. 1977, 77, 349–367.
[20] F. M. Unger, D. Stix, E. Möderndorfer, F. Hammerschmid,
Carbohydr. Res. 1978, 67, 349–356.
[21] R. D. Chambers, D. O’Hagan, R. B. Lamont, S. C. Jaina, J.
Chem. Soc., Chem. Commun. 1990, 1053–1054.
[22] J. Hu, W. Zhanga, F. Wanga, Chem. Commun. 2009, 7465–
7478.
[23] C. Vidil, A. Morère, M. Garcia, V. Barragan, B. Hamdaoui,
H. Rochefort, J. L. Montero, Eur. J. Org. Chem. 1999, 447–467.
[24] S. Vidal, C. Vidil, V. Barragan, A. Morère, M. Garcia, J. L.
Montero, Eur. J. Org. Chem. 2000, 3433–3437.
[25] M. G. Ambrose, R. W. Binkley, J. Org. Chem. 1983, 48, 674–
677.
[26] D. B. Berkowitz, M. Eggen, Q. Shen, D. G. Sloss, J. Org. Chem.
1993, 58, 6174–6176.
[27] a) A. M. Mayer, B. Meyer, Angew. Chem. 1999, 111, 1902; An-
gew. Chem. Int. Ed. 1999, 38, 1784–1788; b) R. Caraballo, H.
Dong, J. P. Ribeiro, J. Jiménez-Barbero, O. Ramstrom, Angew.
Chem. 2010, 122, 599; Angew. Chem. Int. Ed. 2010, 49, 589–
Supporting Information (see footnote on the first page of this arti-
1
cle): Competitive STD experiments and copies of the H and 13C
NMR spectra for all key intermediates and final products.
Eur. J. Org. Chem. 2013, 7776–7784
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7783

L. Cipolla et al.
FULL PAPER
593; c) T. Haselhorst, T. Fiebig, J. C. Dyason, F. E. Fleming,
H. Blanchard, B. S. Coulson, M. von Itzstein, Angew. Chem.
2011, 123, 1087; Angew. Chem. Int. Ed. 2011, 5, 1055–1058; d)
C. Airoldi, S. Giovannardi, B. La Ferla, J. Jiménez-Barbero, F.
Nicotra, Chem. Eur. J. 2011, 17, 13395–13399; e) C. Airoldi, E.
Sironi, C. Dias, F. Marcelo, A. Martins, A. P. Rauter, F. Nico-
tra, J. Jimenez-Barbero, Chem. Asian J. 2013, 8, 596–602.
[28] I. Wiegand, K. Hilpert, R. E. Hancock, Nature Prot. 2008, 3,
163–165.
[30]
[31]
M. Sekiguchi, S. Iida, Proc. Natl. Acad. Sci. USA 1967, 58,
2315–2320.
T. C. Meredith, R. W. Woodard, J. Bacteriol. 2005, 187, 6936–
6942.
[32] C. Airoldi, S. Merlo, F. Nicotra, J. Carbohydr. Chem. 2010, 29,
30–38.
[33] P. Sabbattini, F. Forti, D. Ghisotti, G. Dehò, J. Bacteriol. 1995,
177, 1425–1434.
Received: June 17, 2013
[29] K. A. Datsenko, B. L. Warren, Proc. Natl. Acad. Sci. USA
2000, 97, 6640–6645.
Published Online: October 9, 2013
7784
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 7776–7784